Achievable Rates of Multi-User Millimeter Wave Systems with Hybrid Precoding

TL;DR

This paper introduces a low complexity hybrid precoding algorithm for multi-user mmWave systems, achieving near-optimal rates with reduced hardware complexity and feedback overhead.

Contribution

It proposes a hybrid analog/digital beamforming method tailored for multi-user mmWave systems, with theoretical analysis and practical simulation validation.

Findings

Achieves higher sum rates than analog-only beamforming.

Approaches the performance of unconstrained digital precoding.

Provides a lower bound on achievable rate for single-path channels.

Abstract

Millimeter wave (mmWave) systems will likely employ large antenna arrays at both the transmitters and receivers. A natural application of antenna arrays is simultaneous transmission to multiple users, which requires multi-user precoding at the transmitter. Hardware constraints, however, make it difficult to apply conventional lower frequency MIMO precoding techniques at mmWave. This paper proposes and analyzes a low complexity hybrid analog/digital beamforming algorithm for downlink multi-user mmWave systems. Hybrid precoding involves a combination of analog and digital processing that is motivated by the requirement to reduce the power consumption of the complete radio frequency and mixed signal hardware. The proposed algorithm configures hybrid precoders at the transmitter and analog combiners at multiple receivers with a small training and feedback overhead. For this algorithm, we derive a lower bound on the achievable rate for the case of single-path channels, show its asymptotic optimality at large numbers of antennas, and make useful insights for more general cases. Simulation results show that the proposed algorithm offers higher sum rates compared with analog-only beamforming, and approaches the performance of the unconstrained digital precoding solutions.